The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill implements logic to process potentially untrusted personal data during Data Subject Access Request (DSAR) automation and consent management.
Ingestion points: Data enters the system via the user_id and user_identifier parameters in the ConsentManager and DSARProcessor classes defined in SKILL.md.
Boundary markers: The implementation lacks explicit boundary markers or instructions to the agent to ignore potentially malicious instructions embedded within the user data retrieved from external systems.
Capability inventory: The skill performs database interactions (db.consent_records.insert, db.consent_records.find) and executes data extraction and deletion methods across various data sources (source.extract_user_data, source.delete_user_data).
Sanitization: While the skill uses json.dumps for exporting data, it does not include explicit validation, filtering, or sanitization of the content retrieved from the data sources before processing or returning it.

gdpr-compliance